Abstract: The present invention relates to a system and method for determining sleep quality parameters according to audio analyses, comprising: obtaining an audio recorded signal comprising sleep sounds of a subject; segmenting the signal into epochs; generating a feature vector for each epoch, wherein each of said feature vectors comprises one or more feature parameters that are associated with a particular characteristic of the signal and that are calculated according to the epoch signal or according to a signal generated from the epoch signal; inputting the generated feature vectors into a machine learning classifier and applying a preformed classifying model on the feature vectors that outputs a probabilities vector for each epoch, wherein each of the probabilities vectors comprises the probabilities of the epoch being each of the sleep quality parameters; inputting the probabilities vectors for each epoch into a machine learning time series model and applying a preformed sleep quality time series pattern function

Abstract: Accelerated machine learning using an efficient preconditioner for Kernel Ridge Regression (KRR). A plurality of anchor points may be selected by: projecting an initial kernel onto a random matrix in a lower dimensional space to generate a randomized decomposition of the initial kernel, permuting the randomized decomposition to reorder its columns and/or rows to approximate the initial kernel, and selecting anchor points representing a subset of the columns and/or rows based on their permuted order. A reduced-rank approximation kernel may be generated comprising the subset of columns and/or rows represented by the selected anchor points. A KRR system may be preconditioned using a preconditioner generated based on the reduced-rank approximation kernel. The preconditioned KRR system may be solved to train the machine learning model. This KRR technique may be executed without generating the KRR kernel, reducing processor and memory consumption.

Abstract: Provided herein is a method and system for the estimation of apnea-hypopnea index (AHI), as an indicator for Obstructive sleep apnea (OSA) severity, by combining speech descriptors from three separate and distinct speech signal domains. These domains include the acoustic short-term features (STF) of continuous speech, the long-term features (LTF) of continuous speech, and features of sustained vowels (SVF). Combining these speech descriptors may provide the ability to estimate the severity of OSA using statistical learning and speech analysis approaches.

Abstract: Accelerated machine learning using an efficient preconditioner for Kernel Ridge Regression (KRR). A plurality of anchor points may be selected by: projecting an initial kernel onto a random matrix in a lower dimensional space to generate a randomized decomposition of the initial kernel, permuting the randomized decomposition to reorder its columns and/or rows to approximate the initial kernel, and selecting anchor points representing a subset of the columns and/or rows based on their permuted order. A reduced-rank approximation kernel may be generated comprising the subset of columns and/or rows represented by the selected anchor points. A KRR system may be preconditioned using a preconditioner generated based on the reduced-rank approximation kernel. The preconditioned KRR system may be solved to train the machine learning model. This KRR technique may be executed without generating the KRR kernel, reducing processor and memory consumption.

Abstract: The present invention relates to a system and method for determining sleep quality parameters according to audio analyses, comprising: obtaining an audio recorded signal comprising sleep sounds of a subject; segmenting the signal into epochs; generating a feature vector for each epoch, wherein each of said feature vectors comprises one or more feature parameters that are associated with a particular characteristic of the signal and that are calculated according to the epoch signal or according to a signal generated from the epoch signal; inputting the generated feature vectors into a machine learning classifier and applying a preformed classifying model on the feature vectors that outputs a probabilities vector for each epoch, wherein each of the probabilities vectors comprises the probabilities of the epoch being each of the sleep quality parameters; inputting the probabilities vectors for each epoch into a machine learning time series model and applying a preformed sleep quality time series pattern function

Abstract: Provided herein is a method and system for the estimation of apnea-hypopnea index (AHI), as an indicator for Obstructive sleep apnea (OSA) severity, by combining speech descriptors from three separate and distinct speech signal domains. These domains include the acoustic short-term features (STF) of continuous speech, the long-term features (LTF) of continuous speech, and features of sustained vowels (SVF). Combining these speech descriptors may provide the ability to estimate the severity of OSA using statistical learning and speech analysis approaches.